diffindiff 2.5.8__tar.gz → 2.5.10__tar.gz

{diffindiff-2.5.8 → diffindiff-2.5.10}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: diffindiff
-Version: 2.5.8
+Version: 2.5.10
 Summary: diffindiff: Python library for convenient Difference-in-Differences analyses
 Author: Thomas Wieland
 Author-email: geowieland@googlemail.com
@@ -26,6 +26,8 @@ Dynamic: summary
 
 # diffindiff: Python library for convenient Difference-in-Differences analyses
 
+![Version](https://img.shields.io/pypi/v/diffindiff) ![Downloads](https://static.pepy.tech/badge/diffindiff)
+
 This Python library is designed for performing Difference-in-Differences (DiD) analyses in a convenient way. The package is intended to be used in econometric analyses of natural experiments by researchers in economics, marketing, economic geography, and health sciences. It is designed to cover the entire workflow of a DiD analysis, while not requiring extensive programming skills. The package allows users to construct datasets, define treatment and control groups, and set treatment periods. DiD model analyses may be conducted with both datasets created by built-in functions and ready-to-use external datasets. Both simultaneous and staggered adoption are supported. The library allows for various extensions, such as two-way fixed effects models, group- or individual-specific effects, post-treatment periods, and triple-difference estimations. Additionally, it includes functions for visualizing results, such as plotting DiD coefficients with confidence intervals and illustrating the temporal evolution of staggered treatments. Furthermore, several functions for rigorous treatment configuration and data diagnostics are incorporated.
 
 
@@ -47,7 +49,7 @@ A case study that utilizes the diffindiff library (preprint) is available on [ar
 
 If you use this software, please cite:
 
-Wieland, T. (2026). diffindiff: A Python library for convenient difference-in-differences analyses (Version 2.5.8) [Computer software]. Zenodo. https://doi.org/10.5281/zenodo.18656820
+Wieland, T. (2026). diffindiff: A Python library for convenient difference-in-differences analyses (Version 2.5.10) [Computer software]. Zenodo. https://doi.org/10.5281/zenodo.18656820
 
 
 ## Installation
@@ -198,12 +200,10 @@ See the /tests directory for usage examples of most of the included functions.
 This software was developed without the use of AI-generated code. The GitHub Copilot Chat in Microsoft Visual Studio Code using the GPT-5 mini model (by OpenAI) was used solely to assist in drafting and refining docstrings for documentation. The corresponding guidelines and constraints defined by the author are documented in `AGENTS-docstrings.md` in the [public GitHub repository](https://github.com/geowieland/diffindiff_official).
 
 
-## What's new (v2.5.8)
+## What's new (v2.5.10)
 
 - Bugfixes
-  - didtools.is_balanced() now conducts a truly comprehensive check whether the input panel data is balanced
-  - Fixed bug in didtools.is_numeric() with respect to unintentional modification of the dataset
-  - diddata.create_counterfactual() now catches error if no model was trained by didtools.model_wrapper()
-  - didanalysis.DiffModel.plot_treatment_effects() now treats expections correctly
-- Other:
-  - verbose option in any function which prints NOTEs, with all NOTEs being disabled when verbose = False
+  - Fixed some problems in `didtools.fit_metrics()` which relate to edge cases (e.g., no valid observations)
+- Other
+  - Included deprecation warning for `didtools.model_wrapper()`
+  - Updates and corrections of docstring documentations

{diffindiff-2.5.8 → diffindiff-2.5.10}/README.md

@@ -1,5 +1,7 @@
 # diffindiff: Python library for convenient Difference-in-Differences analyses
 
+![Version](https://img.shields.io/pypi/v/diffindiff) ![Downloads](https://static.pepy.tech/badge/diffindiff)
+
 This Python library is designed for performing Difference-in-Differences (DiD) analyses in a convenient way. The package is intended to be used in econometric analyses of natural experiments by researchers in economics, marketing, economic geography, and health sciences. It is designed to cover the entire workflow of a DiD analysis, while not requiring extensive programming skills. The package allows users to construct datasets, define treatment and control groups, and set treatment periods. DiD model analyses may be conducted with both datasets created by built-in functions and ready-to-use external datasets. Both simultaneous and staggered adoption are supported. The library allows for various extensions, such as two-way fixed effects models, group- or individual-specific effects, post-treatment periods, and triple-difference estimations. Additionally, it includes functions for visualizing results, such as plotting DiD coefficients with confidence intervals and illustrating the temporal evolution of staggered treatments. Furthermore, several functions for rigorous treatment configuration and data diagnostics are incorporated.
 
 
@@ -21,7 +23,7 @@ A case study that utilizes the diffindiff library (preprint) is available on [ar
 
 If you use this software, please cite:
 
-Wieland, T. (2026). diffindiff: A Python library for convenient difference-in-differences analyses (Version 2.5.8) [Computer software]. Zenodo. https://doi.org/10.5281/zenodo.18656820
+Wieland, T. (2026). diffindiff: A Python library for convenient difference-in-differences analyses (Version 2.5.10) [Computer software]. Zenodo. https://doi.org/10.5281/zenodo.18656820
 
 
 ## Installation
@@ -172,12 +174,10 @@ See the /tests directory for usage examples of most of the included functions.
 This software was developed without the use of AI-generated code. The GitHub Copilot Chat in Microsoft Visual Studio Code using the GPT-5 mini model (by OpenAI) was used solely to assist in drafting and refining docstrings for documentation. The corresponding guidelines and constraints defined by the author are documented in `AGENTS-docstrings.md` in the [public GitHub repository](https://github.com/geowieland/diffindiff_official).
 
 
-## What's new (v2.5.8)
+## What's new (v2.5.10)
 
 - Bugfixes
-  - didtools.is_balanced() now conducts a truly comprehensive check whether the input panel data is balanced
-  - Fixed bug in didtools.is_numeric() with respect to unintentional modification of the dataset
-  - diddata.create_counterfactual() now catches error if no model was trained by didtools.model_wrapper()
-  - didanalysis.DiffModel.plot_treatment_effects() now treats expections correctly
-- Other:
-  - verbose option in any function which prints NOTEs, with all NOTEs being disabled when verbose = False
+  - Fixed some problems in `didtools.fit_metrics()` which relate to edge cases (e.g., no valid observations)
+- Other
+  - Included deprecation warning for `didtools.model_wrapper()`
+  - Updates and corrections of docstring documentations

{diffindiff-2.5.8 → diffindiff-2.5.10}/diffindiff/config.py

@@ -4,15 +4,15 @@
 # Author:      Thomas Wieland 
 #              ORCID: 0000-0001-5168-9846
 #              mail: geowieland@googlemail.com              
-# Version:     1.0.29
-# Last update: 2026-10-17 21:17
+# Version:     1.0.31
+# Last update: 2026-11-14 09:55
 # Copyright (c) 2025-2026 Thomas Wieland
 #-----------------------------------------------------------------------
 
 # Basic config:
 
 PACKAGE_NAME = "diffindiff"
-PACKAGE_VERSION = "2.5.8"
+PACKAGE_VERSION = "2.5.10"
 
 VERBOSE = False
 

{diffindiff-2.5.8 → diffindiff-2.5.10}/diffindiff/didanalysis.py

@@ -4,8 +4,8 @@
 # Author:      Thomas Wieland 
 #              ORCID: 0000-0001-5168-9846
 #              mail: geowieland@googlemail.com              
-# Version:     2.4.8
-# Last update: 2026-10-17 21:15
+# Version:     2.4.10
+# Last update: 2026-11-14 09:53
 # Copyright (c) 2024-2026 Thomas Wieland
 #-----------------------------------------------------------------------
 
@@ -45,6 +45,8 @@ class DiffModel:
             Predicted values of the DiD model.
         did_model: statsmodels regression results object
             The fitted DiD model object from statsmodels.
+        timestamp : dict
+            Metadata identifying the package version, function and creation time.
         
         Returns
         -------
@@ -124,6 +126,8 @@ class DiffModel:
             Treatment name to analyse. If None, the first treatment in the model config is chosen.
         after_treatment_col : str, optional
             Column name of an after-treatment indicator, if present.
+        verbose : bool, optional
+            If True, print progress messages.
 
         Returns
         -------
@@ -1122,6 +1126,8 @@ class DiffModel:
         retransform_log_outcome : bool, optional
             If outcome was log-transformed, retransform to original scale for plotting. 
             Default is False.
+        verbose : bool, optional
+            If True, print progress messages.
 
         Returns
         -------
@@ -1334,6 +1340,8 @@ class DiffModel:
         ----------
         treatment : str, optional
             Treatment name to use for placebo assignment.
+        after_treatment_col : str, optional
+            Column name identifying the after-treatment period.
         TG_col : str, optional
             Column name identifying treatment group.
         TT_col : str, optional
@@ -1344,6 +1352,8 @@ class DiffModel:
             Fraction of the divided group to sample when creating placebo units.
         random_state : int, optional
             Seed for random sampling.
+        verbose : bool, optional
+            If True, print progress messages.
 
         Returns
         -------
@@ -1528,6 +1538,8 @@ class DiffModel:
             If not none, filename of plot to be saved.
         save_fig_kwargs : dict, optional
             Optional arguments for plt.savefig() if save_fig is not None.
+        verbose : bool, optional
+            If True, print progress messages.
 
         Returns
         -------
@@ -1686,6 +1698,8 @@ class DiffModel:
             If not none, filename of plot to be saved.
         save_fig_kwargs : dict, optional
             Optional arguments for plt.savefig() if save_fig is not None.
+        verbose : bool, optional
+            If True, print progress messages.
 
         Returns
         -------
@@ -1877,19 +1891,21 @@ class DiffModel:
         outcome_col_expected_PI_lower_CG = f"{outcome_col}{config.DELIMITER}{config.EXPECTED_SUFFIX}{config.DELIMITER}{config.PI_LOWER_SUFFIX}{config.DELIMITER}{config.CG_COL}"
         outcome_col_expected_PI_upper_CG = f"{outcome_col}{config.DELIMITER}{config.EXPECTED_SUFFIX}{config.DELIMITER}{config.PI_UPPER_SUFFIX}{config.DELIMITER}{config.CG_COL}"
 
-        model_data_TG_CG.columns.values[0] = "t"
-        model_data_TG_CG.columns.values[1] = outcome_col_observed_TG
-        model_data_TG_CG.columns.values[2] = outcome_col_observed_CG
-        model_data_TG_CG.columns.values[3] = outcome_col_expected_TG
-        model_data_TG_CG.columns.values[4] = outcome_col_expected_CG
-        model_data_TG_CG.columns.values[5] = outcome_col_expected_CI_lower_TG
-        model_data_TG_CG.columns.values[6] = outcome_col_expected_CI_upper_TG
-        model_data_TG_CG.columns.values[7] = outcome_col_expected_CI_lower_CG
-        model_data_TG_CG.columns.values[8] = outcome_col_expected_CI_upper_CG
-        model_data_TG_CG.columns.values[9] = outcome_col_expected_PI_lower_TG
-        model_data_TG_CG.columns.values[10] = outcome_col_expected_PI_upper_TG
-        model_data_TG_CG.columns.values[11] = outcome_col_expected_PI_lower_CG
-        model_data_TG_CG.columns.values[12] = outcome_col_expected_PI_upper_CG
+        model_data_TG_CG.columns = [
+            "t",
+            outcome_col_observed_TG,
+            outcome_col_observed_CG,
+            outcome_col_expected_TG,
+            outcome_col_expected_CG,
+            outcome_col_expected_CI_lower_TG,
+            outcome_col_expected_CI_upper_TG,
+            outcome_col_expected_CI_lower_CG,
+            outcome_col_expected_CI_upper_CG,
+            outcome_col_expected_PI_lower_TG,
+            outcome_col_expected_PI_upper_TG,
+            outcome_col_expected_PI_lower_CG,
+            outcome_col_expected_PI_upper_CG,
+        ]
                
         if plot_size_auto:
             if model_config["pre_post"]:
@@ -2159,6 +2175,8 @@ class DiffModel:
             If not none, filename of plot to be saved.
         save_fig_kwargs : dict, optional
             Optional arguments for plt.savefig() if save_fig is not None.
+        verbose : bool, optional
+            If True, print progress messages.
 
         Returns
         -------

{diffindiff-2.5.8 → diffindiff-2.5.10}/diffindiff/didanalysis_helper.py

@@ -4,8 +4,8 @@
 # Author:      Thomas Wieland 
 #              ORCID: 0000-0001-5168-9846
 #              mail: geowieland@googlemail.com              
-# Version:     1.2.6
-# Last update: 2026-10-17 18:55
+# Version:     1.2.7
+# Last update: 2026-11-14 09:52
 # Copyright (c) 2025-2026 Thomas Wieland
 #-----------------------------------------------------------------------
 
@@ -147,6 +147,14 @@ def demean_variables(
         Names of spillover variables to demean (default: None).
     interactions : list, optional
         Names of interaction variables to demean (default: None).
+    TG_col : list, optional
+        Names of treatment-group indicator variables to demean.
+    TT_col : list, optional
+        Names of treatment-time indicator variables to demean.
+    after_treatment_col : list, optional
+        Names of after-treatment indicator variables to demean.
+    ATT_col : list, optional
+        Names of average treatment effect on the treated variables to demean.
     FE_unit : bool, optional
         If True, remove unit means (within-unit demeaning). Default is False.
     FE_time : bool, optional
@@ -158,7 +166,7 @@ def demean_variables(
         Convergence tolerance for iterative demeaning. Iteration stops when the
         maximum absolute change across observations is below `tol`. Default is 1e-8.
     verbose : bool, optional
-        If True, print progress and informational messages (default: config.VERBOSE).
+        If True, print progress messages.
 
     Returns
     -------
@@ -1246,8 +1254,48 @@ def extract_model_results(
     ----------
     fit_result : list
         Result returned by ``ols_fit`` or ``ml_fit``.
-    Other parameters
-        Lists of variable names used to identify coefficients (TG_col, TT_col, etc.).
+    TG_col : list, optional
+        Treatment-group variable names used to identify treatment-group coefficients.
+    TT_col : list, optional
+        Treatment-time variable names used to identify treatment-time coefficients.
+    treatment_col : list, optional
+        Treatment variable names used to identify treatment coefficients.
+    after_treatment_col : list, optional
+        After-treatment variable names used to identify after-treatment coefficients.
+    ATT_col : list, optional
+        Variable names used to identify average treatment effect on the treated coefficients.
+    spillover_vars : list, optional
+        Spillover variable names used to identify spillover coefficients.
+    FE_unit_vars : list, optional
+        Unit fixed-effect variable names used to identify fixed-effect coefficients.
+    dummy_unit_original : list, optional
+        Original unit identifiers corresponding to unit fixed-effect dummy variables.
+    FE_time_vars : list, optional
+        Time fixed-effect variable names used to identify fixed-effect coefficients.
+    dummy_time_original : list, optional
+        Original time identifiers corresponding to time fixed-effect dummy variables.
+    FE_group_vars : list, optional
+        Group fixed-effect variable names used to identify fixed-effect coefficients.
+    dummy_group_original : list, optional
+        Original group identifiers corresponding to group fixed-effect dummy variables.
+    ITE_vars : list, optional
+        Variable names used to identify individual treatment effect coefficients.
+    GTE_vars : list, optional
+        Variable names used to identify group treatment effect coefficients.
+    ITT_vars : list, optional
+        Variable names used to identify individual treatment-on-the-treated coefficients.
+    GTT_vars : list, optional
+        Variable names used to identify group treatment-on-the-treated coefficients.
+    TG_x_BG_x_TT_col : list, optional
+        Variable names used to identify treatment-group, background-group and treatment-time interaction coefficients.
+    BG_col : list, optional
+        Background-group variable names used to identify background-group coefficients.
+    TG_x_BG_col : list, optional
+        Variable names used to identify treatment-group and background-group interaction coefficients.
+    BG_x_TT_col : list, optional
+        Variable names used to identify background-group and treatment-time interaction coefficients.
+    covariates : list, optional
+        Covariate names used to identify covariate coefficients.
     verbose : bool, optional
         If True, print progress messages.
 
@@ -1831,6 +1879,8 @@ def remove_duplicates_from_dict(
     ----------
     any_dict : dict
         Dictionary from which to remove duplicates.
+    ignore_key : any, optional
+        Dictionary key to exclude when comparing values for duplicates.
 
     Returns
     -------

{diffindiff-2.5.8 → diffindiff-2.5.10}/diffindiff/diddata.py

@@ -4,8 +4,8 @@
 # Author:      Thomas Wieland 
 #              ORCID: 0000-0001-5168-9846
 #              mail: geowieland@googlemail.com              
-# Version:     2.3.4
-# Last update: 2026-10-17 21:34
+# Version:     2.3.6
+# Last update: 2026-11-14 11:13
 # Copyright (c) 2024-2026 Thomas Wieland
 #-----------------------------------------------------------------------
 
@@ -2958,12 +2958,12 @@ def create_counterfactual(
         verbose = verbose
         )
 
-    model_estimation_error = counterfactual_pred[6]
+    model_estimation_error = counterfactual_pred[5]
     model_estimation_error_text = ""
 
     if model_estimation_error:
 
-        model_estimation_error_text = counterfactual_pred[7]
+        model_estimation_error_text = counterfactual_pred[6]
         print(f"WARNING: The desired {config.MODEL_WRAPPER_AVAILABLE[model_type]} model was not estimated. The following error occurred: '{model_estimation_error_text}'.")
 
     else:

{diffindiff-2.5.8 → diffindiff-2.5.10}/diffindiff/didtools.py

@@ -4,8 +4,8 @@
 # Author:      Thomas Wieland 
 #              ORCID: 0000-0001-5168-9846
 #              mail: geowieland@googlemail.com              
-# Version:     2.2.8
-# Last update: 2026-10-17 18:53
+# Version:     2.2.10
+# Last update: 2026-11-14 11:18
 # Copyright (c) 2025-2026 Thomas Wieland
 #-----------------------------------------------------------------------
 
@@ -25,6 +25,7 @@ from sklearn.preprocessing import StandardScaler
 from sklearn.linear_model import LinearRegression
 from sklearn.model_selection import train_test_split
 from sklearn.neural_network import MLPRegressor
+import warnings
 
 try:
     from xgboost import XGBRegressor
@@ -1402,7 +1403,6 @@ def model_wrapper(
     ):
 
     """
-
     Train a machine learning or OLS regression model and return predictions and metrics.
 
     Parameters
@@ -1430,6 +1430,14 @@ def model_wrapper(
         using this as the class labels.
         Passed to `sklearn.model_selection.train_test_split`;
         see the corresponding documentation.
+    X_train : array-like, optional
+        User-defined training feature data. Supply together with the other training and testing subsets.
+    X_test : array-like, optional
+        User-defined testing feature data. Supply together with the other training and testing subsets.
+    y_train : array-like, optional
+        User-defined training target values. Supply together with the other training and testing subsets.
+    y_test : array-like, optional
+        User-defined testing target values. Supply together with the other training and testing subsets.
     model_n_estimators : int, optional
         The number of estimators in the ensemble.
         Passed to `BaggingRegressor` and `RandomForestRegressor`;
@@ -1507,6 +1515,12 @@ def model_wrapper(
     >>> model_wrapper(y, X, model_type='ols')
     """
 
+    warnings.warn(
+        "didtools.model_wrapper() is deprecated and will be replaced in a future version.",
+        DeprecationWarning,
+        stacklevel=2
+    )
+
     if model_type not in config.MODEL_WRAPPER_AVAILABLE_LIST:
         raise ValueError(f"Please enter a valid model type: {', '.join(config.MODEL_WRAPPER_AVAILABLE_LIST)}.")
     
@@ -1618,6 +1632,7 @@ def model_wrapper(
                     random_state = random_state
                 )
             else:
+                model = {}
                 model_estimation_error = True
                 model_estimation_error_text = "XGBRegressor is not available. Please install xgboost to use this model type."
             
@@ -1630,6 +1645,7 @@ def model_wrapper(
                     random_state = random_state
                 )
             else:
+                model = {}
                 model_estimation_error = True
                 model_estimation_error_text = "LGBMRegressor is not available. Please install lightgbm to use this model type."
             
@@ -1656,7 +1672,8 @@ def model_wrapper(
                 ]
             )
         
-        model.fit(X_train, y_train)
+        if model != {}:
+            model.fit(X_train, y_train)
         
     except Exception as e:
                 
@@ -1763,10 +1780,10 @@ def fit_metrics(
 
     Raises
     ------
-    AssertionError
-        If observed and expected differ in length.
     ValueError
-        If observed and/or expected are not numeric, or contain NaNs when remove_nan is False.
+        If observed and expected differ in length, or contain NaNs when remove_nan is False, or no observations are left after removing NaNs.
+    TypeError
+        If observed and/or expected are not numeric.
 
     Examples
     --------
@@ -1776,21 +1793,23 @@ def fit_metrics(
     observed_no = len(observed)
     expected_no = len(expected)
 
-    assert observed_no == expected_no, "Error while calculating fit metrics: Observed and expected differ in length"
-    
+    if observed_no != expected_no:
+        raise ValueError("Error while calculating fit metrics: Observed and expected differ in length")
+  
     type_errors = []
-    if not pd.api.types.is_numeric_dtype(observed) or not np.issubdtype(observed.dtype, np.number):
-        try:
-            observed = pd.to_numeric(observed)
-        except:
-            type_errors.append("Observed column is not numeric.")
-    if not pd.api.types.is_numeric_dtype(expected) or not np.issubdtype(expected.dtype, np.number):
-        try:
-            expected = pd.to_numeric(expected)
-        except:
-            type_errors.append("Expected column is not numeric.")
+    try:
+        observed = pd.to_numeric(observed)
+    except:
+        type_errors.append("Observed column is not numeric.")
+    try:
+        expected = pd.to_numeric(expected)
+    except:
+        type_errors.append("Expected column is not numeric.")
     if len(type_errors) > 0:
         raise TypeError(f"Error(s) while calculating fit metrics: {', '.join(type_errors)}")
+
+    observed = pd.Series(observed).to_numpy(dtype=np.float64, na_value=np.nan)
+    expected = pd.Series(expected).to_numpy(dtype=np.float64, na_value=np.nan)
     
     if outcome_col is not None and isinstance(outcome_col, str):
         outcome_observed_col = f"{outcome_col}{config.DELIMITER}{config.OBSERVED_SUFFIX}"
@@ -1798,18 +1817,19 @@ def fit_metrics(
     else:
         outcome_observed_col = f"outcome{config.DELIMITER}{config.OBSERVED_SUFFIX}"
         outcome_expected_col = f"outcome{config.DELIMITER}{config.EXPECTED_SUFFIX}"
+
+    obs_exp = pd.DataFrame(
+        {
+            outcome_observed_col: observed, 
+            outcome_expected_col: expected
+            }
+        )
+    observations_input = len(obs_exp)
     
     if remove_nan:
         
         observed = np.array(observed)
         expected = np.array(expected)
-
-        obs_exp = pd.DataFrame(
-            {
-                outcome_observed_col: observed, 
-                outcome_expected_col: expected
-                }
-            )
         
         obs_exp_clean = obs_exp.dropna(
             subset=[
@@ -1820,6 +1840,9 @@ def fit_metrics(
         
         observed = obs_exp_clean[outcome_observed_col].to_numpy()
         expected = obs_exp_clean[outcome_expected_col].to_numpy()
+
+        observed_no = len(observed)
+        expected_no = len(expected)
     
     else:
         
@@ -1830,12 +1853,15 @@ def fit_metrics(
             value_errors.append("Expected data contains NaNs.")
         if len(value_errors) > 0:
             raise ValueError(f"Error(s) while calculating fit metrics: {' '.join(value_errors)} Parameter 'remove_nan' is False.")
-    
+   
+    observations = len(observed)
+
+    if observations == 0:
+        raise ValueError("No observations left for calculating model fit metrics.")
+
     if verbose:
         print("Calculating model fit metrics", end = " ... ")
     
-    observations = len(observed)
-    
     residuals = np.array(observed)-np.array(expected)
     residuals_sq = residuals**2
     residuals_abs = abs(residuals)
@@ -1870,7 +1896,11 @@ def fit_metrics(
     SAR = float(np.sum(residuals_abs))    
     observed_mean = float(np.sum(observed)/observed_no)
     SQT = float(np.sum((observed-observed_mean)**2))
-    RSQ = float(1-(SSR/SQT))    
+    
+    RSQ = np.nan
+    if SQT > 0: 
+        RSQ = float(1-(SSR/SQT))    
+    
     MSE = float(SSR/observed_no)
     RMSE = float(np.sqrt(MSE))
     MAE = float(SAR/observed_no)    
@@ -1881,8 +1911,8 @@ def fit_metrics(
         try:
             indep_vars_no = int(indep_vars_no)
             RSQ_ADJ = (1-(1-RSQ)*((observations-1)/(observations-indep_vars_no-1)))
-        except:
-            pass                
+        except Exception as e:
+            print(f"WARNING: Adj. R-squared can not be computed: {str(e)}")                
 
     model_fit_metrics = {
         list(config.MODEL_FIT_METRICS.keys())[0]: SSR,
@@ -1905,8 +1935,9 @@ def fit_metrics(
         if RSQ_ADJ == np.nan:
             print("NOTE: As no number of independent vars was stated, no Adj. R-Squared is calculated.")
 
-        if len(obs_exp_clean) < len(observed) or len(obs_exp_clean) < len(expected):
-            print("NOTE: Vectors 'observed' and/or 'expected' contain NaNs which were dropped.")
+    if observations < observations_input:
+        observations_input_nan = observations_input-observations
+        print(f"WARNING: {observations_input_nan} pairs of observed and expected were removed due to NaN values.")
             
     modelfit_results = [
         model_residuals,

{diffindiff-2.5.8 → diffindiff-2.5.10}/diffindiff/tests/tests_diffindiff.py

@@ -4,8 +4,8 @@
 # Author:      Thomas Wieland 
 #              ORCID: 0000-0001-5168-9846
 #              mail: geowieland@googlemail.com              
-# Version:     2.2.1
-# Last update: 2026-10-17 14:50
+# Version:     2.2.2
+# Last update: 2026-11-14 11:14
 # Copyright (c) 2025-2026 Thomas Wieland
 #-----------------------------------------------------------------------
 

{diffindiff-2.5.8 → diffindiff-2.5.10}/diffindiff.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: diffindiff
-Version: 2.5.8
+Version: 2.5.10
 Summary: diffindiff: Python library for convenient Difference-in-Differences analyses
 Author: Thomas Wieland
 Author-email: geowieland@googlemail.com
@@ -26,6 +26,8 @@ Dynamic: summary
 
 # diffindiff: Python library for convenient Difference-in-Differences analyses
 
+![Version](https://img.shields.io/pypi/v/diffindiff) ![Downloads](https://static.pepy.tech/badge/diffindiff)
+
 This Python library is designed for performing Difference-in-Differences (DiD) analyses in a convenient way. The package is intended to be used in econometric analyses of natural experiments by researchers in economics, marketing, economic geography, and health sciences. It is designed to cover the entire workflow of a DiD analysis, while not requiring extensive programming skills. The package allows users to construct datasets, define treatment and control groups, and set treatment periods. DiD model analyses may be conducted with both datasets created by built-in functions and ready-to-use external datasets. Both simultaneous and staggered adoption are supported. The library allows for various extensions, such as two-way fixed effects models, group- or individual-specific effects, post-treatment periods, and triple-difference estimations. Additionally, it includes functions for visualizing results, such as plotting DiD coefficients with confidence intervals and illustrating the temporal evolution of staggered treatments. Furthermore, several functions for rigorous treatment configuration and data diagnostics are incorporated.
 
 
@@ -47,7 +49,7 @@ A case study that utilizes the diffindiff library (preprint) is available on [ar
 
 If you use this software, please cite:
 
-Wieland, T. (2026). diffindiff: A Python library for convenient difference-in-differences analyses (Version 2.5.8) [Computer software]. Zenodo. https://doi.org/10.5281/zenodo.18656820
+Wieland, T. (2026). diffindiff: A Python library for convenient difference-in-differences analyses (Version 2.5.10) [Computer software]. Zenodo. https://doi.org/10.5281/zenodo.18656820
 
 
 ## Installation
@@ -198,12 +200,10 @@ See the /tests directory for usage examples of most of the included functions.
 This software was developed without the use of AI-generated code. The GitHub Copilot Chat in Microsoft Visual Studio Code using the GPT-5 mini model (by OpenAI) was used solely to assist in drafting and refining docstrings for documentation. The corresponding guidelines and constraints defined by the author are documented in `AGENTS-docstrings.md` in the [public GitHub repository](https://github.com/geowieland/diffindiff_official).
 
 
-## What's new (v2.5.8)
+## What's new (v2.5.10)
 
 - Bugfixes
-  - didtools.is_balanced() now conducts a truly comprehensive check whether the input panel data is balanced
-  - Fixed bug in didtools.is_numeric() with respect to unintentional modification of the dataset
-  - diddata.create_counterfactual() now catches error if no model was trained by didtools.model_wrapper()
-  - didanalysis.DiffModel.plot_treatment_effects() now treats expections correctly
-- Other:
-  - verbose option in any function which prints NOTEs, with all NOTEs being disabled when verbose = False
+  - Fixed some problems in `didtools.fit_metrics()` which relate to edge cases (e.g., no valid observations)
+- Other
+  - Included deprecation warning for `didtools.model_wrapper()`
+  - Updates and corrections of docstring documentations

{diffindiff-2.5.8 → diffindiff-2.5.10}/setup.py

@@ -7,7 +7,7 @@ def read_README():
     
 setup(
     name='diffindiff',
-    version='2.5.8',
+    version='2.5.10',
     description='diffindiff: Python library for convenient Difference-in-Differences analyses',
     packages=find_packages(include=["diffindiff", "diffindiff.tests"]),
     include_package_data=True,